JPEG Processing Neural Operator for Backward-Compatible Coding

2025-07-31

AI Summary

JPEG’s backward compatibility and chrominance fidelity are inherently difficult to reconcile. To address this, we propose JPNeO—the first next-generation JPEG algorithm that integrates neural operators while fully preserving the standard JPEG encoding protocol (i.e., without modifying entropy coding, quantization tables, or other core components). Our method rests on three key innovations: (1) a high mutual information spatial hypothesis that guides encoder-side feature optimization and decoder-side reconstruction enhancement; (2) a lightweight neural operator module seamlessly embedded into the conventional JPEG pipeline, significantly improving chrominance preservation and reconstruction fidelity; and (3) superior performance over state-of-the-art artifact removal methods—achieving substantial gains in PSNR and MS-SSIM while reducing model parameters by 37% and memory footprint by 29%. Extensive experiments confirm JPNeO’s strong performance, low computational overhead, and plug-and-play compatibility with existing JPEG infrastructure.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Non-convex Optimization
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Despite significant advances in learning-based lossy compression algorithms, standardizing codecs remains a critical challenge. In this paper, we present the JPEG Processing Neural Operator (JPNeO), a next-generation JPEG algorithm that maintains full backward compatibility with the current JPEG format. Our JPNeO improves chroma component preservation and enhances reconstruction fidelity compared to existing artifact removal methods by incorporating neural operators in both the encoding and decoding stages. JPNeO achieves practical benefits in terms of reduced memory usage and parameter count. We further validate our hypothesis about the existence of a space with high mutual information through empirical evidence. In summary, the JPNeO functions as a high-performance out-of-the-box image compression pipeline without changing source coding's protocol. Our source code is available at https://github.com/WooKyoungHan/JPNeO.

Problem

• Enhancing JPEG compression with backward compatibility
• Improving chroma component and reconstruction fidelity
• Reducing memory usage and parameter count

Innovation

• Backward-compatible JPEG neural operator
• Neural-enhanced encoding and decoding
• Reduced memory and parameter usage